Qwen API pricing: every model, every rate
Qwen is the largest non-US model family by parameter spread, from 0.6B up to Max tiers. Below is every billable variant.
Check current 0rates →
Free to sign up · $1 minimum top-up · No prepayment
Full rate card
88 billable models. 0 carry a rate verified against the vendor's own published pricing; the rest are gateway rates only. USD per 1M tokens, verified 2026-09-16.
| Model | Input | Output | Out/In | Official (in / out) |
|---|---|---|---|---|
qwen2.5-32b | $10 | $30 | 3.0× | not verified |
qwen2.5-vl-72b-instruct | $8 | $24 | 3.0× | not verified |
qwen2.5-vl-32b-instruct | $4 | $12 | 3.0× | not verified |
qwen3-coder | $3 | $12 | 4.0× | not verified |
qwen2-vl-72b-instruct | $2.25 | $2.25 | 1.0× | not verified |
qwen2.5-72b-instruct | $2 | $6 | 3.0× | not verified |
qwen2.5-math-72b-instruct | $2 | $6 | 3.0× | not verified |
qwen3-235b-a22b-think | $1.5 | $6 | 4.0× | not verified |
qwen3.7-max | $1.25 | $3.75 | 3.0× | not verified |
qwen3.5-omni-flash | $1.1 | $6.65 | 6.0× | not verified |
qwen-72b | $1 | $1 | 1.0× | not verified |
qwen2.5-32b-instruct | $1 | $3 | 3.0× | not verified |
qwen2.5-coder-14b-instruct | $1 | $3 | 3.0× | not verified |
qwen2.5-coder-32b-instruct | $1 | $3 | 3.0× | not verified |
qwen2.5-vl-7b-instruct | $1 | $2.5 | 2.5× | not verified |
qwen3-vl-235b-a22b | $1 | $4 | 4.0× | not verified |
qwen3.8-max | $1 | $3 | 3.0× | not verified |
qwen3.8-max-0902 | $1 | $3 | 3.0× | not verified |
qwen-mt-plus | $0.9 | $2.7 | 3.0× | not verified |
qwen-max | $0.8 | $3.2 | 4.0× | not verified |
qwen-omni-turbo | $0.8 | $2.2504 | 2.8× | not verified |
qwen-vl-plus | $0.75 | $2.25 | 3.0× | not verified |
qwen3-coder-480b-a35b-instruct | $0.75 | $3.75 | 5.0× | not verified |
qwen3.6-max-preview | $0.65 | $3.9 | 6.0× | not verified |
qwen2.5-vl-3b-instruct | $0.6 | $1.8 | 3.0× | not verified |
qwen3-max | $0.6 | $3 | 5.0× | not verified |
qwen3-max-preview | $0.6 | $3 | 5.0× | not verified |
qwen2.5-14b-instruct | $0.5 | $1.5 | 3.0× | not verified |
qwen2.5-14b-instruct-1m | $0.5 | $1.5 | 3.0× | not verified |
qwen2.5-coder-7b-instruct | $0.5 | $1 | 2.0× | not verified |
qwen2.5-math-7b-instruct | $0.5 | $1 | 2.0× | not verified |
qwen3-coder-plus | $0.5 | $2.5 | 5.0× | not verified |
qwen3-next-80b-a3b | $0.5 | $2 | 4.0× | not verified |
qwen-plus-busikao | $0.4 | $1 | 2.5× | not verified |
qwen-vl-max | $0.4 | $1.6 | 4.0× | not verified |
qwen-mt-turbo | $0.35 | $0.9765 | 2.8× | not verified |
qwen3-235b-a22b | $0.35 | $1.4 | 4.0× | not verified |
qwen3.5-397b-a17b | $0.3 | $1.8 | 6.0× | not verified |
qwen3.6-27b | $0.3 | $1.8 | 6.0× | not verified |
Qwen/Qwen3-Reranker-8B | $0.28 | — | not verified | |
qwen-plus-character | $0.25 | $0.7 | 2.8× | not verified |
qwen2-vl-7b-instruct | $0.25 | $0.25 | 1.0× | not verified |
qwen2.5-7b-instruct | $0.25 | $0.5 | 2.0× | not verified |
qwen2.5-7b-instruct-1m | $0.25 | $0.5 | 2.0× | not verified |
qwen3.6-plus | $0.25 | $1.5 | 6.0× | not verified |
qwen3.8-27b | $0.25 | $1.5 | 6.0× | not verified |
qwen3-coder-30b-a3b-instruct | $0.225 | $1.125 | 5.0× | not verified |
qwen-plus | $0.2 | $0.6 | 3.0× | not verified |
qwen3-vl-235b-a22b-instruct | $0.2 | $0.8 | 4.0× | not verified |
qwen3-vl-235b-a22b-thinking | $0.2 | $2 | 10.0× | not verified |
qwen3.5-122b-a10b | $0.2 | $1.6 | 8.0× | not verified |
qwen3.5-plus | $0.2 | $1.2 | 6.0× | not verified |
qwen3.7-plus | $0.2 | $0.8 | 4.0× | not verified |
qwen3.6-35b-a3b | $0.1875 | $1.125 | 6.0× | not verified |
qwen3-14b | $0.175 | $0.7 | 4.0× | not verified |
qwen2.5-3b-instruct | $0.15 | $0.45 | 3.0× | not verified |
qwen3-0.6b | $0.15 | $0.6 | 4.0× | not verified |
qwen3-1.7b | $0.15 | $0.6 | 4.0× | not verified |
qwen3-4b | $0.15 | $0.6 | 4.0× | not verified |
qwen3-coder-flash | $0.15 | $0.75 | 5.0× | not verified |
qwen3.5-27b | $0.15 | $1.2 | 8.0× | not verified |
Qwen/Qwen3-Reranker-4B | $0.14 | — | not verified | |
qwen3.5-35b-a3b | $0.125 | $1 | 8.0× | not verified |
qwen3-235b-a22b-instruct-2507 | $0.115 | $0.46 | 4.0× | not verified |
qwen3-235b-a22b-thinking-2507 | $0.115 | $1.15 | 10.0× | not verified |
qwen3-30b-a3b | $0.1 | $0.4 | 4.0× | not verified |
qwen3-30b-a3b-instruct-2507 | $0.1 | $0.4 | 4.0× | not verified |
qwen3-30b-a3b-think | $0.1 | $1.2 | 12.0× | not verified |
qwen3-30b-a3b-thinking-2507 | $0.1 | $1.2 | 12.0× | not verified |
qwen3-vl-30b-a3b-instruct | $0.1 | $0.4 | 4.0× | not verified |
qwen3-vl-30b-a3b-thinking | $0.1 | $1.2 | 12.0× | not verified |
qwen3-vl-plus | $0.1 | $0.8 | 8.0× | not verified |
qwen3-8b | $0.09 | $0.36 | 4.0× | not verified |
qwen3-vl-8b-instruct | $0.09 | $0.35 | 3.9× | not verified |
qwen3-vl-8b-thinking | $0.09 | $1.05 | 11.7× | not verified |
qwen3-32b | $0.08 | $0.32 | 4.0× | not verified |
qwen3-vl-32b-instruct | $0.08 | $0.32 | 4.0× | not verified |
qwen3-vl-32b-thinking | $0.08 | $0.32 | 4.0× | not verified |
qwen3-next-80b-a3b-instruct | $0.075 | $0.6 | 8.0× | not verified |
qwen3-next-80b-a3b-thinking | $0.075 | $0.6 | 8.0× | not verified |
qwen3.8-flash | $0.075 | $0.235 | 3.1× | not verified |
qwen3-rerank | $0.05 | $0 | not verified | |
qwen3.5-flash | $0.05 | $0.2 | 4.0× | not verified |
qwen-flash | $0.025 | $0.2 | 8.0× | not verified |
qwen-turbo | $0.025 | $0.1 | 4.0× | not verified |
qwen-turbo-1101 | $0.025 | $0.1 | 4.0× | not verified |
qwen3-vl-flash | $0.025 | $0.2 | 8.0× | not verified |
Qwen/Qwen3-Reranker-0.6B | $0.0105 | — | not verified |
Output is billed at a multiple of input on most models. The Out/In column is that multiple — it matters more than the input price once output dominates your bill.
Across 88 billable models the rate spans $0.0105 (Qwen/Qwen3-Reranker-0.6B) to $10 (qwen2.5-32b) per million input tokens — a 952× spread. Picking the wrong tier is usually the single most expensive mistake here.
What Alibaba is good at
Qwen coder models are strong enough for agentic work and priced well below comparable US models.
Before you compare: Qwen Max tiers are billed by cache state and context length. Standard rates shown.
See all 827 models → or use the cost calculator with your own token split.